Tag: ai

Intel goes to Nervana to sort out its intelligence

Chipzilla has put its artificial intelligence efforts into a single business group led by former CEO of Nervana,  Naveen Rao.

For those who came in late, Intel bought Nervana in the firm belief that the next big thing will be AI powered IT innovation and machine learning.

Writing in his bog, Rao outlined how the Artificial Intelligence Products Group will work across multiple units. Part of the group’s remit will be to bring AI costs down and forge standards. Rao said the group will combine engineering, labs, software, and hardware from its portfolio.

Intel is building an AI lab and a centralised organisation, reporting directly to CEO Brian Krzanich, to make it all work.

This is classic organisational strategy, accelerating delivery by creating  a cross-product group directly reporting to the CEO.

Don’t make the Samsung S8 angry it will have Bixby

Samsung has confirmed that it is Bixby digital assistant will be part of the Galaxy S8 that’ll be unveiled later this month.

The S8 will have a dedicated Bixby button on its side to make it easier to access the assistant.

It will have three key features to sort itself out from the herd. The first is that a Bixby-enabled app will allow the assistant to perform every task that the app normally supports using touch.

It will also have context awareness, which means that when Bixby is activated, it’ll can understand the current context and the state of the app that you’re in without interrupting the work that you’re doing.

Samsung says that Bixby is smart enough to understand commands with incomplete commands, meaning that you do not have to remember the exact phrase that you have to say to perform a task with an assistant. Bixby will ask you for more information when performing a task and then execute it.

Several apps on the Galaxy S8 will be Bixby-enabled at launch, and Samsung plans to add more over time. The company will release an SDK so that third-party app developers can add Bixby support.

Samsung said that the assistant will first appear on Samsung smartphones and then expand to all Samsung appliances.

“Since Bixby will be implemented in the cloud, if a device has an internet connection and simple circuitry to receive voice inputs, it can connect with Bixby,” the company said.

Amazon wants help improving Alexa

Online book seller Amazon.com has launched a new programme to help students build capabilities into its voice controlled assistant Alexa.

The e-commerce company said it is paying for a year long doctoral fellowship at four universities.

Working with professors, the Alexa Fund Fellows will help students tackle complex technology problems in class on Alexa, like how to convert text to speech or process conversation.

Amazon and Google is locked in a race to develop and make cash from artificial intelligence. Amazon has made it easy for third party developers to create skills for Alexa so it can get better faster – a tactic it now is extending to the classroom.

The other idea is that Amazon might be able to recruit sought after engineers whose studies will make them more familiar with Alexa than with other voice controlled assistants.

Schools signed up for the programme include Carnegie Mellon, Johns Hopkins, the University of Southern California and Canada’s University of Waterloo.

Doug Booms, vice president of worldwide corporate development at Amazon, said that the fellowship’s goal is to excite the next generation of scholars about natural language understanding and other voice technologies, not to produce research for Amazon.

Students’ projects will remain their own intellectual property.

For example at the University of Waterloo, students are improving Alexa’s interaction with air conditioners so it understands requests to cool a room to its normal temperature, without requiring the user to specify a number in Celsius.

UK sorts out insurance for self-driving cars

accidentcarinwashingtondcThe UK plans to introduce new insurance rules to ensure victims of accidents involving self-driving cars are compensated quickly.

The move will remove a major obstacle for the nascent industry. Self-driving car introduction has been hampered by legal hurdles in several countries as insurers and legislators try to establish who would ultimately be responsible in the event of an accident.

Transport Minister Chris Grayling said the public needed to be protected in the event of an incident and the framework to allow insurance for these new technologies will be out this week.

A single insurance product will be available to cover a driver when a vehicle is being used conventionally, as well as when the car is being used in autopilot mode, the transport ministry said in a statement.

The Blighty government wants to encourage the development and testing of autonomous driving technology to build an industry to serve a market it reckons could be worth about $1.1 trillion worldwide by 2025.

Japanese carmaker Nissan is due to test autonomous cars in London later this month after initial tests on public roads in the southern English town of Milton Keynes late last year.

The UK will also set out plans to improve infrastructure such as charging points for electric vehicles, the fastest growing sector for new car sales in the country and key to meeting environmental targets.

Oculus ordered to pay up on ZeniMax tech

keep_calm_and_love_your_patent_lawyer_2_inch_round_magnet-re8c2c059dc99401ca676f1a1e58344f5_x7js9_8byvr_324A US jury in Texas ordered Oculus, and other defendants to pay a combined $500 million to ZeniMax Media, a video game publisher that claims Oculus stole its technology.

The jury thought that in 2014, Oculus used ZeniMax’s computer code to launch the Rift virtual-reality headset. ZeniMax alleges that video game designer John Carmack developed core parts of the Rift’s technology while working at a ZeniMax subsidiary. Oculus hired Carmack in 2013.

ZeniMax Chief Executive Robert Altman hailed the verdict and said in a statement the company was considering seeking an order blocking Oculus and Facebook from using its code. It is unclear what impact that would have on the Rift’s market availability.

However, the jury ruled that none of the defendants misappropriated ZeniMax’s trade secrets, but it did think that Oculus’ use of computer code directly infringed on ZeniMax’s copyright. The jurors held Carmack and different Oculus co-founders Palmer Luckey and Brendan Iribe liable for forms of infringement.

The jury also found Oculus liable for breaching a non-disclosure agreement Luckey signed with ZeniMax in 2012, when he began corresponding about virtual reality with Carmack.

Carmack worked for id Software before that company was acquired by ZeniMax. He is now the chief technology officer at Oculus.

Facebook Chief Executive Mark Zuckerberg testified last month during the three-week trial that none of ZeniMax’s proprietary code was incorporated into the Rift.

In a statement, Oculus spokeswoman Emily Bauer noted the jury’s finding on trade secrets theft and said the company would appeal. “We’re obviously disappointed by a few other aspects of today’s verdict, but we are undeterred,” she said. “Oculus products are built with Oculus technology.”

Linked in and eBay millionaires invest to save us from AI

cybermen__quot_delete_quot__campaign_by_degaspiv-d33hjoaTwo millionaires have each committed $10 million to save the world from the troubles of AI.

Reid Hoffman, the founder of LinkedIn, and eBay founder Pierre Omidyar’s non-profit are spending a fortune funding academic research and development aimed at keeping artificial intelligence systems ethical.

The fund received an additional $5 million from the Knight Foundation and two other $1 million donations from the William and Flora Hewlett Foundation and Jim Pallotta, founder of the Raptor Group. The $27 million reserve is being anchored by MIT’s Media Lab and Harvard’s Berkman Klein Center for Internet and Society.

While the project acknowledges that AI has its uses they are worried that things can go wrong. The most critical challenge is making sure that machines are not trained to perpetuate and amplify the same human biases that plague society.

The money will pay for research to investigate how socially responsible artificially intelligent systems can be designed to, say, keep computer programs that are used to make decisions in fields like education, transportation and criminal justice accountable and fair.

The group wants to talk with the public about and foster understanding of the complexities of artificial intelligence. The two universities will form a governing body along with Hoffman and the Omidyar Network to distribute the funds.

Nearly half of our current jobs will be gone in 25 years

Robby the Robot - Wikimedia CommonsWharton School of Business at the University of Pennsylvania has warned that all the developed nations on earth will see job loss rates of up to 47 per cent within the next 25 years.

The statistic is based on a recent Oxford University study and includes blue and white collar jobs. So far, the loss has been restricted to the blue collar variety, particularly in manufacturing so no one has cared that much as this has been happening since the 1960s.

The new trend is not creating new jobs either. By 2034, just a few decades, mid-level jobs will be by and large obsolete.

So far the benefits have only gone to the ultra-wealthy, the top 1 per cent. This coming technological revolution is set to wipe out what looks to be the entire middle class.

Accountants, doctors, lawyers, teachers, bureaucrats, and financial analysts beware: your jobs are not safe. Soon computers will analyze and compare reams of data to make financial decisions or medical ones. There will be less of a chance of fraud or misdiagnosis, and the process will be more efficient. Not only are these folks in trouble, such a trend is likely to freeze salaries for those who remain employed, while income gaps only increase in size.

Unfortunately the report suggests that it is too late to turn Luddite and break up the machines. Governments will need to sort out some form of retraining, although it is not clear what the nasty fleshy pink lumps can do that robots can’t.

 

 

AMD releases AI based Radeons with basic instinct

BasicInstinct002AMD is announcing a new series of Radeon-branded products today, targeted at machine intelligence and deep learning enterprise applications.

Dubbed the Radeon Instinct, the chip is a GPU-based solution for deep learning, inference and training. AMD has also issued a new free, open-source library and framework for GPU accelerators, dubbed MIOpen.

MIOpen is made for high-performance machine intelligence applications and is optimized for deep learning frameworks in AMD’s ROCm software suite.

The first products are the Radeon Instinct MI6, the MI8, and the MI25. The 150W Radeon Instinct MI6 accelerator is powered by a Polaris-based GPU, packs 16GB of memory (224GB/s peak bandwidth), and can manage 5.7 TFLOPS of peak FP16 performance when the wind is behind it and it is going downhill.

It also includes the Fiji-based Radeon Instinct MI8. Like the Radeon R9 Nano, the Radeon Instinct MI8 features 4GB of High-Bandwidth Memory (HBM) with peak bandwidth of 512GB/s. AMD claims the MI8 will offer up to 8.2 TFLOPS of peak FP16 compute performance, with a board power that typical falls below 175W.

The Radeon Instinct MI25 accelerator uses AMD’s next-generation Vega GPU architecture and has a board power of approximately 300W. All the Radeon Instinct accelerators are passively cooled but when installed into a server chassis you can bet there will be plenty of air flow.

Like the recently released Radeon Pro WX series of professional graphics cards for workstations, Radeon Instinct accelerators will be built by AMD. All the Radeon Instinct cards will also support AMD MultiGPU (MxGPU) hardware virtualisation.

Microsoft’s Chinese AI is clever enough to censor itself

beijing cybercafeSoftware King of the World has admitted that its Chinese flavoured AI chat bot will not talk about anything that the authorities behind the bamboo curtain don’t want them to talk about.

Xiaoice would not directly respond to questions surrounding topics deemed sensitive by the Chinese state including the Tiananmen Square massacre of 1989 or “Steamed Bun Xi,” a nickname of Chinese President Xi Jinping.

“Am I stupid? Once I answer you’d take a screengrab,” read one answer to a question that contained the words “topple the Communist Party.”

Mentioning Donald “Prince of Orange” Trump also drew an evasive response from the chat bot. “I don’t want to talk about it,” Xiaoice says. Fair enough who does?

Microsoft has admitted that there was some filtering around Xiaoice’s interaction.

“We are committed to creating the best experience for everyone chatting with Xiaoice,” a Microsoft spokesperson said. “With this in mind, we have implemented filtering on a range of topics.” The tech giant did not further elaborate to which specific topics the filtering applied.

Microsoft says that Xiaoice engages in conversations with over 40 million Chinese users on social media platform like Weibo and WeChat.

LingLong creates DingDong in smart home industry

Linglong-Dingdong-Lautsprecher-1024x576-31f5edc41d756a0cChinese outfit LingLong has created an AI based assistant it has dubbed the DingDong which is making a sing song in the consumer electronics market.

The gear has a music library of three million songs, can take memos and share updates regarding news, traffic and weather in what the firm calls ‘cinema-like sound quality’

It speaks Cantonese and Mandarin, which means it can roll into the lucrative Chinese market and get a head start on its Western rivals.

It costs $118 and answers questions, gives directions and plays music in high quality 320Kbps format

The device comes in four colours: red for prosperity, white for purity, black for money and purple because it is pretty.

In the west, Amazon is the leaders in this space. It released its Echo in 2014 – smart speaker powered by Alexa. Users can ask Alexa to do a range of activities such as request an Uber or order their usually from Dominos – and there is more than three million units in the world.

Most DingDong owners use the technology as a music player, or as someone to talk to.